Professional Animal Scientist-Review copy--Confidential
Review: updated scientific evidence on the welfare of
gestating sows kept in different housing systems
Journal:
The Professional Animal Scientist
r
Fo
Manuscript ID:
Manuscript Type:
Date Submitted by the Author:
Key words:
Review
n/a
McGlone, John; Texas Tech University, Pork Industry Institute
Pe
Complete List of Authors:
PAS-12-01172.R1
Pig, Sow, Gestation housing, Welfare
er
ew
vi
Re
ScholarOne, 375 Greenbrier Drive, Charlottesville, VA, 22901
Page 1 of 50
Professional Animal Scientist-Review copy--Confidential
1
Running Heading: Updated Review of Gestation Sow Housing Literature
Review: updated scientific evidence on the welfare of gestating sows kept in
different housing systems
John J McGlone1, PAS, ACAWS
Laboratory of Animal Behavior, Physiology and Welfare2, Department of Animal and
Food Sciences, Texas Tech University, Lubbock, TX 79409-2141
iew
ev
rR
ee
rP
Fo
Correspondence:
John J. McGlone, PhD
Professor and Institutional Official
Animal and Food Sciences Department
Texas Tech University
Lubbock, TX 79409-2141
phone: 806-742-2805, ext. 246
fax: 806-742-4003
john.mcglone@ttu.edu
1
2
Correspondence: john.mcglone@ttu.edu.
http://www.depts.ttu.edu/animalwelfare/
1
ScholarOne, 375 Greenbrier Drive, Charlottesville, VA, 22901
Professional Animal Scientist-Review copy--Confidential
Page 2 of 50
2
1
ABSTRACT
Gestation sow housing is a contemporary animal welfare issue with legislative actions in the
3
USA states to ban individual gestation sow systems. This review sought to summarize the
4
scientific literature since earlier reviews were published in 2004 and 2005. Seventeen papers
5
comparing effect of housing systems on the welfare of gestating sows were published from
6
2005 to 2012. Stalls/crates, tether housing, and group pens including conventional group
7
pens, loose-house pens, electronic sow feeders (ESF) pens, and hoop barns were reported.
8
The majority of recent findings were similar to the previous conclusions from papers and
9
reviews. Cortisol concentrations were not different between stalled and group penned sows
rP
Fo
2
but were higher shortly after mixing in concert with increased sow aggression and skin
11
lesions. The effects of gestation housing system on neutrophil:lymphocyte ratio was still
12
mixed. In the recent literature, stereotypies were greater among stalled sows compared with
13
group penned sows which may reflect a change in sow genetics since the basic housing
14
systems have not changed. Recent papers indicated more sitting/standing inactive, leg and
15
claw problems, and farrowing rate among stalled sows, higher lying among group penned
16
sows, and no differences in other reproductive and productivity measures between stalled and
17
group penned sows. Weaning-to-estrus intervals were similar between stalled and group
18
penned sows in contrast to previous reviews that observed a shorter interval among stalled
19
compared with grouped sows. In conclusion, recent papers found similar productivity,
20
physiology health and behavior among individually or grouped sows during gestation.
iew
ev
rR
ee
10
21
22
Key words: Pig, Sow, Gestation housing, Welfare
23
24
25
2
ScholarOne, 375 Greenbrier Drive, Charlottesville, VA, 22901
Page 3 of 50
Professional Animal Scientist-Review copy--Confidential
3
26
27
INTRODUCTION
Gestation sow housing is a social issue in Europe, North America, Australia and New
Zealand. The primary concern is about the welfare of pregnant sows kept in individual
29
gestation crates (also called stalls). The gestation crate does not allow the sow to turn around
30
or make normal postural adjustments. However, the gestation crate is the most common
31
system used in the USA because it minimizes space needs and provides a mean to
32
individually feed and care for pregnant sows. Tether housing and crates or stalls for
33
individual sows have been perceived as housing systems that cause poor welfare. Therefore,
34
European countries and several states in USA (FL, CA, AZ, OR, CO, RI, at least) have
35
banned or are phasing out gestation crates. Retailers (grocery and restaurant) have made
36
public statements that they will preferentially buy pork from farms that do not use gestation
37
crates. Still, the industry defends the gestation crate as a viable sow housing system.
ee
rP
38
Fo
28
Two reviews of the scientific literature about the welfare of pregnant sows were
rR
published in 2004 and 2005 (McGlone et al., 2004; Rhodes et al., 2005). In the past 8 years
40
more research has been published that sought to compare the welfare of pregnant sows in
41
individual gestation crates and group pens. The objectives of this review were (a) to
42
summarize the current scientific literature on the welfare of gestation sow housing systems
43
and (b) to determine if the conclusions of the recent literature differ from the conclusions of
44
the previous reviews. Finally, we speculate on the current status of the welfare of pregnant
45
sows in housing systems based on the newer scientific literature.
iew
ev
39
46
47
SYSTEMS REVIEWED
48
The “control” system by which other systems are compared is the gestation stall (also
49
called crate). This is the control or standard system because at this time, in North America, it
50
is the most common system to keep pregnant sows. Any change away from the crate will
3
ScholarOne, 375 Greenbrier Drive, Charlottesville, VA, 22901
Professional Animal Scientist-Review copy--Confidential
Page 4 of 50
4
incur an economic cost to pork producers and ultimately the consumer. We should know if
52
this seemingly inevitable economic cost is associated with better, the worse or the same sow
53
welfare. While we find a plethora of systems in which to keep pregnant sows, only a few
54
have generated sufficient scientific literature in which comparisons can be made. These
55
group-housing systems include the (a) group penning with or without individual feeding
56
stalls, and (b) ESF. Other systems included in the earlier reviews from which we found
57
insufficient new literature were the outdoor system, neck or girth tethers, turn-around stalls,
58
and the Hurnik-Morris system. Because only one paper investigated girth tethers and
59
onlytwo papers examined loose-housing in hoop barns, theses system will be mentioned but
60
not highlighted.
rP
61
Fo
51
The scope of this updated literature review is to compare individual with group
ee
keeping systems for pregnant sows. Papers have been published with single systems or in
63
ways that ask specific questions about sow preferences or other questions about sow
64
behavior, physiology or health. These papers were not considered here because they did not
65
compare sow keeping systems.
ev
66
rR
62
Group housing systems are more accepted by some people because as they allow the
iew
67
animals to express social behaviors and sows can turn around. However, group pens also pose
68
welfare problems due to fighting of sows to compete for limited resources and in the
69
establishment of a social hierarchy (Spoolder et al., 2009). Adopting a housing system that is
70
good for the overall welfare of the sows should rely on scientific research results that cover
71
several relevant welfare indicators such as behavior, physiology, health, reproduction, and
72
productivity (CAST, 2009).
73
74
MEASURES OF SOW WELFARE
4
ScholarOne, 375 Greenbrier Drive, Charlottesville, VA, 22901
Page 5 of 50
Professional Animal Scientist-Review copy--Confidential
5
75
Multiple measures of sow welfare should be used in any evaluation of
76
housing/penning systems. For this review, as with past reviews, we considered three types of
77
measures to be direct or indirect measures of sow welfare. These categories include (a)
78
physiology, (b) behavior, and (c) health, reproductive performance and productivity. The
79
categories could be split or merged, however, given that relatively few papers collected all
80
measures.
81
Measures of physiology focused on the stress hormone cortisol (which may rise
during stress) and immune measures (which may increase or decrease during stress; Salak-
83
Johnson and McGlone, 2007). Other measures of physiology like heart and respiratory rates
84
were not often measured in the more recent literature.
rP
85
Fo
82
Measures of behavior include maintenance behaviors (sitting, standing, lying, feeding,
ee
drinking) and various forms of oral-nasal-facial (ONF) behaviors, including stereotyped
87
behaviors. One particular type of ONF behavior called “stereotypies” is thought by some to
88
be critical measures of sow welfare and thus was summarized. Because of the difficulty of
89
defining which ONF behaviors are and are not “stereotypies, we largely considered all ONF
90
behaviors. However, we acknowledge other views related to the interpretation of ONF
91
behaviors (Damm et al., 2005).
iew
ev
92
rR
86
The category of measures including health, reproduction and productivity is very
93
broad. Health measures that were tied to a putative stress response were included in
94
measures of physiology. Measures of reproductive health and productivity are connected and
95
logically can be in one category. Health also includes bone and foot and leg health and body
96
and vulva lesions.
97
98
METHODS
5
ScholarOne, 375 Greenbrier Drive, Charlottesville, VA, 22901
Professional Animal Scientist-Review copy--Confidential
Page 6 of 50
6
99
The scientific literature was searched electronically. Agricola, Pubmed, and Google
100
Scholar were searched from the period of Jan 1, 2005 through August 1, 2012. Keywords
101
included sow and/or pig, housing or penning. In addition, journals known to contain papers
102
on pig biology were searched including The Journal of Animal Science, The Professional
103
Animal Scientist, Livestock Science, Applied Animal Behaviour Science, Animal, Animal
104
Welfare, Journal of Applied Animal Welfare Science and Journal of Swine Health and
105
Production.
Fo
106
107
rP
108
FINDINGS
Results from 17 publications from 2005 to 2012 comparing housing systems for
gestating sows in a variety of aspects relating to welfare were summarized in each animal
110
welfare category. Methodologies used in the reviewed studies are presented in Table 1.
111
Details of the findings by measure are provided in Tables 2 to 6.
112
114
Physiology
ev
113
rR
ee
109
Blood or salivary cortisol concentrations have been used to determine stress among
iew
115
gestating sows in housing systems. Comparing stall- and group-kept sows, most authors
116
found no differences in cortisol concentrations. The recent literature confirms the earlier
117
reviews that sows in stalls or group pens may equally adapt to the environments by mid
118
gestation. Group-kept sows are often mixed and this results in fighting and an elevation in
119
cortisol (Jansen et al., 2007). Associations among mixing of grouped sows, aggressive
120
interactions, and cortisol have been investigated in studies dealing with grouping of sows in a
121
single system but not including a comparison of sows in crates and group pens. Therefore,
122
these papers were not included in this review. After the social hierarchy was established, no
123
differences were observed in cortisol concentrations between crated and group penned sows
6
ScholarOne, 375 Greenbrier Drive, Charlottesville, VA, 22901
Page 7 of 50
Professional Animal Scientist-Review copy--Confidential
7
124
(Hulbert and McGlone, 2006; Karlen et al., 2007). Cortisol was also elevated when group-
125
penned gestating sows were moved to farrowing crates compared with sows crated during
126
gestation and moved to farrowing crates (Sorrells et al., 2007).
127
For stress-related immune measures, there were inconsistent results. Karlen et al.
(2007) found higher neutrophil:lymphocyte ratios measured at 15 weeks of gestation among
129
stalled sows compared with group penned sows. In contrast, Hulbert and McGlone (2006) did
130
not find differences in neutrophil:lymphocyte ratios, % neutrophil phagocytosis, and
131
neutrophil chemotaxis measured at mid gestation between individually crated and group
132
housed sows.
134
ee
135
Behavior
rP
133
Fo
128
Some authors argue ONF behavior should include feeding and drinking behaviors
(McGlone et al., 2004) while others argue ONF should not include feeding and drinking oral
137
behaviors (Damm et al.(2005). Setting this argument aside, one class of ONF behaviors --
138
the stereotypies directed to objects such as bars and feeders -- were greater among stalled
139
sows compared with group penned sows (Karlen et al., 2007). These recent findings are
140
consistent with the results from the previous reviews.
iew
ev
rR
136
141
Sitting or standing inactive for long periods may indicate poor welfare according to
142
both the previous reviews and the current literature. Lying, in the opposite way, may reflect
143
good welfare as shown if the lying behaviors in group penned sows increased. Hulbert and
144
McGlone (2006) found no difference in behavior for sows kept in stalls or group pens.
145
Findings about aggressive behaviors were consistent in the recent literature that
146
evaluated group housing. Sows show an increase in aggressive interactions after weaning or
147
mixing after breeding (Jansen et al., 2007). Seguin et al. (2006) and Salak-Johnson et al.
7
ScholarOne, 375 Greenbrier Drive, Charlottesville, VA, 22901
Professional Animal Scientist-Review copy--Confidential
Page 8 of 50
8
148
(2007) investigated the effects of space allowance on lesion scores but not on aggressive
149
behavior; nor did they have a direct comparison with individually-kept sows.
150
151
152
Health, Reproductive Performance & Productivity
Body lesions (mostly resulting from fighting especially shortly after mixing) and
locomotory problems have been investigated to determine welfare of sows housed in different
154
systems. Concerning foot and leg problems, results were mixed. Conventional group penned
155
sows and those kept in ESF had more lameness scores and claw lesions than stalled sows
156
providing there was no bedding (Harris et al., 2006; Anil et al., 2007). In contrast, Karlen et
157
al. (2007) reported more lameness among stalled sows than group penned sows provided
158
deep bedding. However, in an epidemiological study, Ryan et al.(2010), observed similar
159
lameness scores and joint pathologies between sows housed in stalls and loose-housed pens
160
where bedding was provided. Therefore, lameness may be more a function of the absence of
161
bedding rather than housing system.
rR
ee
rP
The previous reviews concluded that weaning-to-estrus interval in gestating sows
ev
162
Fo
153
reared in stalls was shorter than those kept in group pens. However, from the current results,
164
weaning-to-estrus interval and also weaning-to-service interval were not different between
165
stalled and group penned sows. Farrowing rates were lower in group penned sows compared
166
to stalled sows. However, Hulbert and McGlone (2006) did not find a difference in
167
reproduction between crated and grouped sows. Other reproductive parameters also indicate
168
no difference due to housing. On the whole, reproductive performance was similar for sows
169
in group pens and individual crates in the current literature.
170
iew
163
Several reports showed no difference in litter size for sows housed in stalls compared
171
with any type of group pens (Harris et al., 2006; Hulbert and McGlone, 2006; Jansen et al,
172
2007; Karlen et al., 2007; Salak-Johnson et al., 2007; Chapinal et al., 2010a). However, one
8
ScholarOne, 375 Greenbrier Drive, Charlottesville, VA, 22901
Page 9 of 50
Professional Animal Scientist-Review copy--Confidential
9
173
study found greater litter size among group penned sows compared with individually crated
174
sows (Seguin et al., 2006).
175
Sorrells et al.(2006) studied prenatal stress in piglets born from gilts housed in stalls
176
or pens. These findings may not be directly relevant to the welfare of gestating sows but may
177
relate to the welfare in the entire system since the welfare of piglets may be impacted by
178
gestation housing system.
179
181
Fo
180
CONCLUSIONS
Cortisol concentrations may be used as a physiological sign of stress. Cortisol
rP
concentrations were higher among group penned sows shortly after mixing (Jansen et al.,
183
2007). However, cortisol concentrations were not different between stalled and group-
184
penned sows in mid gestation. Differences in immune measures were not detected between
185
crated and penned sows. Physiological measures, on the whole, were not different between
186
individually crated and group-kept pregnant gilts and sows. This is consistent with the earlier
187
reviews.
ev
rR
ee
182
Overall ONF behaviors did not differ among sows in group or individual systems. In
189
contrast, stereotypies were higher among stalled sows than group penned sows in one paper
190
(Karlen et al., 2007) but not another (Hulbert and McGlone, 2006). Excessive sitting and
191
standing inactive may indicate poor welfare. More sitting and/or standing inactive were
192
reported in stalled sows compared to groped sows (Munsterhjelm et al., 2008; Weng et al.,
193
2009), however, Hulbert and McGlone (2006) did not find this.
194
iew
188
Aggression commonly occurs in group penned sows particularly shortly after mixing
195
and during feeding. This aggression leads to body or vulva lesions (Jansen et al., 2007; Salak-
196
Johnson et al., 2007;Chapinal et al., 2010a;Chapinal et al., 2010b).
9
ScholarOne, 375 Greenbrier Drive, Charlottesville, VA, 22901
Professional Animal Scientist-Review copy--Confidential
Page 10 of 50
10
197
Because the housing systems have not changed, it is possible that the selection for
198
improved reproductive performance or other traits has resulted in sows that express more
199
stereotypies when individually crated. One cannot draw this conclusion on the basis of just a
200
few recent papers. However, it is an interesting hypothesis that recent intensive genetic
201
selection may have changed the behavior of pregnant sows perhaps with increased
202
aggressive, sitting and ONF behaviors.
203
Both recent studies and the earlier reviews found that leg and foot problems had a
greater incidence in group penned sows than crated sows (McGlone et al., 2004; Harris et al.,
205
2006; Anil et al., 2007). However, lameness issues may be more a function of bedding (or
206
lack of bedding) than penning system (Karlen et al., 2007).
rP
207
Fo
204
In the recent literature, no differences were observed in weaning-to-estrus interval
ee
among stalled versus group-kept sows. Higher farrowing rates were reported among stalled
209
sows compared to group penned sows. Other reproductive parameters such as ovulation rate,
210
number of embryo, embryo survival, backfat thickness, and body condition scores were
211
similar among sows kept in stalls or group pens. Most studies indicated no difference in litter
212
size and birth weight of piglets born from sows housed in stalls or group pens. The majority
213
of the current findings from 2005 to 2012 were not different from those in the previous
214
review. On the whole, few physiological, behavioral or health differences were reported
215
between individually crated and group-kept pregnant gilts and sows.
iew
ev
rR
208
216
217
218
Implications
Overall, group penning and individual crating of pregnant sows support about the
219
same level of measurable sow welfare. Grouping has negative consequences such as more
220
body lesions, and aggression when grouped. Individual crating may cause more stereotyped
221
behaviors of unknown cause or consequence, although this observation was not consistent
10
ScholarOne, 375 Greenbrier Drive, Charlottesville, VA, 22901
Page 11 of 50
Professional Animal Scientist-Review copy--Confidential
11
222
across studies. A compilation of the recent scientific literature does not draw one to a
223
different conclusion than the literature summarized in 2004 and 2005. Improvements in the
224
welfare of sows are possible in each pregnant sow keeping system.
ACKNOWLEDGEMENTS
The authors thank Dr. Kim Guay for technical review of this paper and Wirawat
Chaya for gathering the literature. The author declare no conflicts of interest during the
rP
Fo
writing of this review.
LITERATURE CITED
ee
Chapinal, N., J. L. Ruiz de la Torre, A. Cerisuelo, J. Gasa, M. D. Baucells, J. Coma, A. Vidal,
and X. Manteca. 2010a. Evaluation of welfare and productivity in pregnant sows kept in
stalls or in 2 different group housing systems. J. Vet. Behav. 5:82.
rR
Chapinal, N., J. L. Ruiz de la Torre, A. Cerisuelo, J. Gasa, M. D. Baucells, and X. Manteca.
2010b. Aggressive behavior in two different group-housing systems for pregnant sows. J.
Appl. Anim. Welf. Sci. 13:137.
iew
ev
Council for Agricultural Science and Technology (CAST). 2009. Scientific Assessment of
the Welfare of Dry Sows Kept in Individual Accommodations. Issue Paper 42. CAST, Ames,
Iowa.
Damm, B. I., B. Forkman, M. Giersing, J. Ladewig, and H. Houe. 2005. Letter to the editor:
An alternative perspective on the compilation of the scientific literature comparing housing
systems for gestating sows using metaanalysis on measures of physiology, behavior,
performance, and health. Prof. Anim. Sci. 21:139.
Harris, M. J., E. A. Pajor, A. D. Sorrells, S. D. Eicher, B. T. Richert, and J. N. MarchantForde. 2006. Effects of stall or small group gestation housing on the production, health and
behavior of gilts. Livest. Sci. 102:171.
Hulbert, L. E. and J. J. McGlone. 2006. Evaluation of drop versus trickle-feeding systems for
crated or group-penned gestating sows. J. Anim. Sci. 84:1004.
Jansen, J., R. N. Kirkwood, A. J. Zanella, and R. J. Tempelman. 2007. Influence of gestation
housing on sow behavior and fertility. J. Swine Health Prod. 15:132.
11
ScholarOne, 375 Greenbrier Drive, Charlottesville, VA, 22901
Professional Animal Scientist-Review copy--Confidential
Page 12 of 50
12
Karlen, G. A. M., P. H. Hemsworth, H. W. Gonyou, E. Fabrega, A. D. Strom, and R. J.
Smits. 2007. The welfare of gestating sows in conventional stalls and large groups on deep
litter. Appl. Anim. Behav. Sci. 105:87.
Lammers, P. J., M. S. Honeyman, J. W. Mabry, and J. D. Harmon. 2007. Performance of
gestating sows in bedded hoop barns and confinement stalls. J. Anim. Sci. 85:1311.
McGlone, J. J., E. H. von Borrel, J. Deen, A. K. Johnson, D. G. Levis, M. Meunier-Salaun, J.
Morrow, D. Reeves, J. L. Salak-Johnson, and P. L. Sunberg. 2004. Review: Compilation of
the scientific literature comparing housing systems for gestating sows and gilts using
measures of physiology, behavior, performance, and health. Prof. Anim. Sci. 20:105.
Munsterhjelm, C. A. Valros, M. Heinonen, O. Halli, and O. A. T. Peltoniemi. 2008. Housing
during early pregnancy affects fertility and behavior of sows. Reprod. Dom. Anim. 43:584.
Fo
rP
Rhodes, R. T., M. C. Appleby, K. Chinn, L. Douglas, L. D. Firkins, K. A. Houpt, C. Irwin, J.
J. McGlone, P. Sundberg, L. Tokach, and R.W. Wills. 2005. A comprehensive review of
housing for pregnant sows. J. Am. Vet. Med. Assoc. 227:1580.
Ryan, W. F., P. B. Lynch, and J. V. O’Doherty. 2010. A survey of bone integrity from cull
sows in Ireland. Irish Vet. J. 63:689.
ee
rR
Salak-Johnson, J . L., S. R. Niekamp, S. L. Rodriguez-Zas, M. Ellis, and S. E. Curtis. 2007.
Space allowance for dry, pregnant sow in pens: Body condition, skin lesions, and
performance. J. Anim. Sci. 85:1758.
ev
Salak-Johnson, J. L., and J. J. McGlone. 2007. Making sense of apparently conflicting data:
stress and immunity in swine and cattle. J. Anim. Sci. 85: Suppl: E81-8.
iew
Seguin, M. J. D. Barney, and T. M. Widowski. 2006. Assessment of a group-housing system
for gestating sows: Effects of space allowance and pen size on the incidence of superficial
skin lesions, changes in body condition, and farrowing performance. J. Swine Health Prod.
14:89.
Sorrells, A. D., S. D. Eicher, M. J. Harris, E. A. Pajor, and B. T. Richert. 2007. Periparturient
cortisol, acute phase cytokine, and acute phase protein profiles of gilts housed in groups or
stalls during gestation. J. Anim. Sci. 85:1750.
Sorrells, A. D., S. D. Eicher, K. A. Scott, M. J. Harris, E. A. Pajor, D. C. Lay, Jr., and B. T.
Richert. 2006. Post natal behavioral and physiological responses of piglets from gilts housed
individually or in groups during gestation. J. Anim. Sci. 84:757.
Spoolder, H. A. M., M. J. Geudeke, C. M. C. Van der Peet-Schwering, and N. M. Soede.
2009. Group housing of sows in early pregnancy: A review of success and risk factors.
Livest. Sci. 125:1.
Anil, S. S., L. Anil., J. Deen, and S. K. Baidoo. 2007. Factors associated with claw lesions in
gestating sows. J. Swine Health Prod. 15:78.
12
ScholarOne, 375 Greenbrier Drive, Charlottesville, VA, 22901
Page 13 of 50
Professional Animal Scientist-Review copy--Confidential
13
van der Staay, F. J., T. Schuurman, M. Hulst, M. Smits, J. Prickaerts, G. Kenis, and S. M.
Korte. 2010. Effects of chronic stress: A comparison between tethered and loose sows.
Physiol. Behav. 100:154.
van Wettere, W. H. E. J., S. J. Pain, P. G. Stott, and P. E. Hughes. 2008. Mixing gilts in early
pregnancy does not affect embryo survival. Anim. Rep. Sci. 104:382.
Weng, R. C., S. A. Edwards, and L. C. Hsia. 2009. Effect of individual, group or ESF
housing in pregnancy and individual or group housing in lactation on sow behavior. AsianAust. J. Anim. Sci. 22:1574.
iew
ev
rR
ee
rP
Fo
13
ScholarOne, 375 Greenbrier Drive, Charlottesville, VA, 22901
Professional Animal Scientist-Review copy--Confidential
Page 14 of 50
14
Table 1. Methodologies in reviewed recent literature.
Authors
Animals, parities
Housing
systems
compared
Chapinal et al.
Landrace x Large
Stall (S)
(2010a)
White; parities 1-9 group pen (P)
unprotected
ESF group pen
(ESF)
Fo
Space
allowance
(m2/sow)
1.29 (S)
2.76 (P)
2.32
(ESF)
Bedding
Group sizes,
replicates
No
20 stalled sows, 2
group pens of 10
sows, and an ESF
pen of 20 sows x 3
replicates
rP
ee
rR
Chapinal et al.
(2010b)
Landrace x Large
White; parities 1-9
Group pen (P)
ESF group pen
(ESF)
Ryan et al. (2010)
Cull sows in 8-9
parities (studied at
the abattoir)
Stall, loosehouse
van der Staay et
al.(2010)
Great
Yorkshire/Large
White x Dutch
Landrace; parities
2-13
Landrace x
Yorkshire
gilts
Tethering (T), 2.2 (L)
loose house (L)
Weng et al.(2009)
Stall S)
group pen (P)
ESF group pen
(ESF)
2.76 (P)
2.32
(ESF)
No
NA
Yes
(straw for
loosehouse)
1.41(S)
1.80 (P)
NA (ESF)
No




Behavior (week 1-2 and
then every week)
Haptoglobin and PigMAP (week 11-13 and
16-18)
Reproduction
Productivity
Lameness and injury
2 group pens of 10
sows and an ESF
pen of 20 sows x 3
replicates
4,120 stalled sows
from 5 farms, 2,900
loose-housed sows
from 2 farms (group
size of 100)
59 loose-housed
sows (65-130
sows/pen), 70
tethered sows

Aggressive behavior


Lameness score
Joint pathologies


16 stalled sows, 16
group penned sows
(5 sows/pen), 32
ESF sows

Plasma cortisol
Weights of adrenal,
pituitary, and spleen
mRNA of PrP and βglobin in brain
Behavior at 8
physiological stages from
estrus to dry
ev
No
Measures
iew
14
ScholarOne, 375 Greenbrier Drive, Charlottesville, VA, 22901


Page 15 of 50
Professional Animal Scientist-Review copy--Confidential
15
Munsterhjelm et
al.(2008)
van Wettere et
al.(2008)
Anil et al.(2007)
Jansen et al.(2007)
Yorkshire (64%),
Yorkshire x
Finnish Landrace
(36%), averaged
parities 2.4
Purebred maternal
(Large
White)/terminal
(Duroc)
Gilts and sows;
parities 0-8
Stall, group
pen
Landrace and
Yorkshire; parities
0, 1, 3
Stall, group
pen
Stall, group
pen
Fo
Karlen et al.(2007) Landrace x Large
White; parities ≥1
rP
Stall, ESF
group pen,
Yes (deep
litter for
pen)
240 stalled sows,
240 group penned
sows (12 pens of 20
sows)



Behavior (day 27 after
weaning)
Reproduction
Productivity
1.36 (S)
2.40 (P)
No

Reproduction
1.20 (S)
NA (ESF)
No
24 stalled sows, 4
group pens of 6 sows
x 3 mixing
treatments
82 stalled sows, 102
group penned sows
(no group size
mentioned)
48 stalled sows, 48
group penned sows
(mixed with other
sows in 4 pens to
form a group of 50)

Claw lesions measured at
day 110 of gestation

Behavior (0-3 d after
mixing)
Lesion scores (+1, -1 d of
mixing)
Salivary cortisol (+1, -1 d
of mixing)
Reproduction
Behavior (week 1, 9 of
gestation)
Salivary cortisol (day 5,
week 9 of gestation)
Hematology and
lymphocyte proliferation
(week 15 of gestation)
Reproduction
Productivity
1.44 (S)
5.10 (P)
ee
rR
NA (S)
2.10 (P)
No
ev
Stall, group
pen
1.26 (S)
2.30 (P)
Yes (deep
rice hull
for group
pen)
iew
640 sows in total,
group size of 85






Lammers et
al.(2007)
25% Hamshire x
50% Yorkshire x
25% Landrace;
Stall (S), hoop
barns (H)
1.26 (S)
3.44 (H)
Yes (deep
cornstalks
for hoop
552 stalled sows,
405 group penned in
hoop barns (32
15
ScholarOne, 375 Greenbrier Drive, Charlottesville, VA, 22901


Professional Animal Scientist-Review copy--Confidential
Page 16 of 50
16
parities 0-2
Salak-Johnson et
al.(2007)
Sows (PIC) in
parities ≥1
Sorrells et
al.(2007)
Landrace x
Yorkshire gilts
Harris et al.(2006)
Landrace x
Yorkshire gilts
Hulbert and
McGlone (2006)
barns)
Stall, group
pen
Fo
Camborough 22
gilts
Stall, group
pen
rP
Stall, group
pen
Stall, group
pen
sows/pen, 2
pens/barn, 2 barns)
51 stalled sows, 166
group penned sows
(group size of 5)
No
No
16 stalled gilts, 8
group pens of 4 gilts

1.35 (S)
2.40 (P)
No
16 stalled gilts, 8
group pens of 4 gilts

ee

rR
1.33 (S)
1.28 (S)
No
80 stalled gilts, 80
group penned gilts
ev
iew
Seguin et al.(2006) Yorkshire; parities
2.8
Sorrells et
al.(2006)
Landrace x
Yorkshire gilts
Stall, group
pen
Stall, group
pen


1.29 (S)
1.40,
2.30, and
3.30 (P)
1.32 (S)
2.34 (P)
2.00 (S)
2.30,
2.80, and
3.20 (P)
No
1.35 (S)
2.41 (P)
No
98 stalled sows, 9
small group penned
sows (group size of
11-19), 6 large group
penned sows (group
size of 22-31)
16 stalled gilts, 8
group pens of 4 gilts
16
ScholarOne, 375 Greenbrier Drive, Charlottesville, VA, 22901









Productivity
Lesion scores (every 3 d
for the first 2 weeks, then
every week of gestation)
Cortisol , cytokines, and
acute phase proteins
during gestation and after
farrowing
Behavior (week 4, 6, 9,
and 13 of gestation)
Lesion scores (every 2
weeks)
Productivity
Behavior (50-60 d of
gestation)
Cortisol (50-60 d of
gestation)
Hematology and immune
(50-60 d of gestation)
Reproduction
Productivity
Lesion scores (+1,-1 d of
mixing, then every week
for 5 weeks)
Productivity
Behavior (isolation test in
piglets)
Page 17 of 50
Professional Animal Scientist-Review copy--Confidential
17

Fo
rP
ee
rR
ev
iew
17
ScholarOne, 375 Greenbrier Drive, Charlottesville, VA, 22901
Cortisol, acute phase
proteins, IgG, and TNF-α
in piglets
Professional Animal Scientist-Review copy--Confidential
Page 18 of 50
18
rR
ee
rP
Fo
Table 2. Effects of housing on physiology of gestating sows
Findings from the previous reviews in 2004
Findings from 2005-2012
(McGlone et al., 2004) and Rhodes et
al.(2005)
Cortisol
 Circulating cortisol concentrations of
 Tethered sows had higher plasma
stalled and group penned sows did not
cortisol than loose-housed sows
differ. However, the concentrations in
(van der Staay et al., 2010).
tethered sows were greater compared
No variation in concentrations of
to those in stalled or group penned
plasma cortisol measured during 50sows.
60 days of gestation between sows
housed in crates and group pens
(Hulbert and McGlone, 2006).
 Salivary cortisol concentrations
measured at 1 and 9 weeks of
gestation were not different between
stalled and group penned sows
(Karlen et al., 2007). However,
Jansen et al. (2007) found
significantly higher salivary cortisol
concentrations in group penned
sows the day of relocation compared
to prior moving. In addition, group
penned gilts had higher salivary
cortisol levels shortly after moving
to farrowing crates compared to
stalled gilts (Sorrells et al., 2007).

There were no differences in
neutrophil: lymphocyte ratio, %
neutrophil phagocytosis, and
neurophil chemotaxis between
crated and group penned sows
(Hulbert and McGlone, 2006).
Stalled sows had higher neutrophil:
lymphocyte ratio than group penned
sows (Karlen et al., 2007).
No differences in cytokines m RNA
expressions and acute phase proteins
between stalled and group penned
gilts (Sorrells et al., 2007)



Cardiovascular functions
 There was no information comparing
cardiovascular response of sows in
different housing systems.
iew
ev
Immune responses
 Housing did not influence neutrophil:
lymphocyte ratio, immunoglobulin
concentrations, antibody response
against sheep red blood cell antigens,
and natural killer cell activity.
Stalled and group penned sows had
similar heart rates at rest and during
activities (Harris et al., 2006).
18
ScholarOne, 375 Greenbrier Drive, Charlottesville, VA, 22901
Page 19 of 50
Professional Animal Scientist-Review copy--Confidential
19
iew
ev
rR
ee
rP
Fo
Table 3. Effects of housing on behavior of gestating sows
Findings from the previous reviews in 2004
Findings from 2005-2012
(McGlone et al., 2004) and Rhodes et
al.(2005)
Oral-nasal-facial (ONF) behaviors and
stereotypies
 ONF behaviors in stalled and group
 Stalled and group penned sows had
penned sows were similar. Tethered
no difference in percentage of time
sows might show higher or lower
spent for ONF behaviors (Hulbert
ONF behavior depending on
and McGlone, 2006; Munsterhjelm
experimental designs. Sows housed in
et al., 2008). However, Chapinal et
group pens expressed less time spent
al.(2010a) reported that stalled sows
for stereotypies than those housed in
spent higher proportion of time for
stalls.
ONF behaviors than group penned
sows. Percentages of time spent for
stereotyped interactions with bars
and feeders were greater in stalled
sows than group penned sows which
was contrast to the percentage of
time spent for stereotyped
interactions with floor (Karlen et al.,
2006).
Locomotory and postural behaviors
 Gilts kept in turn-around stalls had
 Munsterhjelm et al.(2008) found
more standing activity than stalled
that stalled sows engaged more in
sows. An increase in space allowance
standing inactive which was an
reduced standing and sitting inactive
indicator of poor welfare than group
of sows in group pens. Group pen
penned sows. More standing, sitting,
sows changed more postures than
rooting, and drinking were also
stalled sows in farrowing stalls and
observed in stalled sows than group
more restless at parturition.
penned sows and sows housed in
Electronic Sow Feeding (ESF)
system (Weng et al., 2009). ESF
sows had higher time spent for lying
and less time spent for rooting and
drinking (Weng et al., 2009). This
was consistent with the results of
Chapinal et al. (2010a) indicating
more lying in ESF sows than stalled
sows and group penned sows with
trickle feeding system. Karlen et al.
(2007) reported more lying and less
standing/walking at week 9 of
gestation among stalled sows than
group penned sows. However,
Hulbert and McGlone (2006) did not
find any difference in time budgets
for lying, sitting, and
standing/walking in crated and
group penned sows.
19
ScholarOne, 375 Greenbrier Drive, Charlottesville, VA, 22901
Professional Animal Scientist-Review copy--Confidential
Page 20 of 50
20
Social behaviors
 Agonistic behavior was higher in
group penned sows particularly
during mixing and feeding compared
to stalled and tethered sows.
Increased space allowance resulted in
a decreased agonistic behavior and
social interactions among group
penned sows.

iew
ev
rR
ee
rP
Fo
Total number of active and passive
aggressive encounters was greater in
group penned sows than stalled
sows and during feeding than during
after mixing (Jansen et al., 2007).
When 2 types of group pens were
compared, ESF penned sows had
higher frequency of aggressive
behavior than conventional group
penned sows (Chapinal et al.,
2010b). Karlen et al.(2007) also
noted the decrease of fighting in
group penned sows at 9 weeks of
gestation compared to that in the
first week. Results revealed by
Hulbert and McGlone (2006)
indicate no difference in agonistic
behavior of crated sows and group
penned sows at 50-60 days of
gestation.
20
ScholarOne, 375 Greenbrier Drive, Charlottesville, VA, 22901
Page 21 of 50
Professional Animal Scientist-Review copy--Confidential
21

Locomotory problems
 Incidence of claw lesions was double
higher in loose-housed sows than
stalled sows and was also more
common than tethered sows. Sows
housed in group pens with deep
bedding were found to have higher
claw lesions than stalled and tethered
sows. Without bedding, lameness
incidence was greater in ESF and
group penned sows compared to
iew
ev
rR
ee
rP
Fo
Table 4. Effects of housing on health of gestating sows
Findings from the previous reviews in 2004
Findings from 2005-2012
(McGlone et al., 2004) and Rhodes et
al.(2005)
Vulva biting
 Vulva injuries related to biting were
 Higher numbers of sows developed
more frequent in ESF sows than other
vulva biting in ESF and group
types of group housing systems.
penned sows compared to stalled
There was no prevalence of vulva
sows (Chapinal et al., 2010).
biting in stalled sows and tethered
However, Harris et al. (2006) did
sows.
not detect difference in vulva lesion
scores between stalled and group
penned sows.
Body lesion scores
 Group penned sows had 3 times in
 Group penned sows had higher
percentage of body lesions caused by
lesion scores than stalled sows after
fighting compared to stalled sows.
mixing and these scores correlated
Provision of roughage reduced a risk
with the total number of agonistic
of developing body lesions in group
interactions (Jansen et al., 2007).
penned sows.
Salak-Johnson et al.(2007) noticed a
decrease in space allowance (1.4 vs.
2.3 and 3.3 m2) leading to higher
lesion scores in group penned sows.
However, a study by Seguin et
al.(2006) indicated that there was no
difference in lesion scores of sows
provided space allowance of 2.3,
2.8, and 3.2 m2. The study by
Hulbert and McGlone (2006)
showed no dissimilarity in lesion
scores measured at mid-gestation
between crated sows and group
penned sows.
Group penned gilts lacking bedding
had more foot and leg problems
investigated at 13 weeks of gestation
than stalled gilts (Harris et al.,
2006). Karlen et al.(2007) reported
higher lameness scores measured at
9 and 13 weeks of gestation among
stalled sows than group penned
sows provided deep bedding.
Comparing ESF with stall housing,
Anil et al.(2007) discovered higher
proportion of sows kept in ESF
developing claw lesions. However,
based on the study of Ryan et
al.(2010), lameness scores and joint
pathology of culled sows previously
21
ScholarOne, 375 Greenbrier Drive, Charlottesville, VA, 22901
Professional Animal Scientist-Review copy--Confidential
Page 22 of 50
22
stalled and free –access stalled sows.
housed in stalls or loose-housed
pens were not different
iew
ev
rR
ee
rP
Fo
22
ScholarOne, 375 Greenbrier Drive, Charlottesville, VA, 22901
Page 23 of 50
Professional Animal Scientist-Review copy--Confidential
23

iew
ev
rR
ee
rP
Fo
Table 5. Effects of housing on reproductive performance of gestating sows
Findings from the previous reviews in 2004
Findings from 2005-2012
(McGlone et al., 2004) and Rhodes et
al.(2005)
Weaning-to-estrus interval
 Sows housed in stalls had shorter
 Munsterhjelm et al.(2008) observed
weaning-to-estrus interval than those
no difference in weaning-to-estrus
housed in group pens. Group penned
between stalled and group penned
sows had similar estrus detection rate
sows. This was consistent with a
and estrus duration but longer time to
report by Jansen et al.(2007). Also,
onset and onset to ovulation than
weaning-to-service did not differ
stalled sows.
between stalled and group penned
sows (Harris et al., 2006;
Munsterhjelm et al., 2008).
However, Lammers et al.(2007)
noticed shorter weaning-to-service
interval among stalled sows than
group penned sows.
Farrowing rate
 farrowing rate in stalled sows was
 Stalled sows had higher farrowing
higher than group penned sows.
rate than group penned sows (Karlen
Tethered sows had lower farrowing
et al., 2007). In contrary, Hulbert
rate than stalled and group penned
and McGlone (2006) found no
sows. There was no difference in
variation in farrowing rates of sows
farrowing rates of outdoor reared
housed in stall and group pens.
sows compared to indoor stalled
sows. Stalled sows had greater
farrowing rate than loose-housed
sows during summer.
 van Wettere et al.(2008) also
detected a similar in ovulation rate,
number of embryos, and embryo
survival in gilts kept in stalls and
group pens.
Body condition scores (BCS)
evaluated before and after entering
gestation did not differ in stalled and
group penned sows (Seguin et al.,
2006). Body weight and backfat
thickness were also not different
between stalled and group penned
sows (Harris et al., 2006; Hulbert
and McGlone, 2006; Chapinal et al.,
2010a). Space allowance positively
affected body weight, BCS, and
backfat thickness (Salak-Johnson et
al., 2007). There was a similarity in
body weight at farrowing and
weaning and backfat thickness at
23
ScholarOne, 375 Greenbrier Drive, Charlottesville, VA, 22901
Professional Animal Scientist-Review copy--Confidential
Page 24 of 50
24
breeding, farrowing, and weaning in
sows housed in crates or group pens
(Hulbert and McGlone, 2006).
iew
ev
rR
ee
rP
Fo
24
ScholarOne, 375 Greenbrier Drive, Charlottesville, VA, 22901
Page 25 of 50
Professional Animal Scientist-Review copy--Confidential
25
iew
ev
rR
ee
rP
Fo
Table 6. Effects of housing during gestation on productivity of sows and behavior and
physiology of offspring
Findings from the previous reviews in 2004
Findings from 2005-2012
(McGlone et al., 2004) and Rhodes et
al.(2005)
Litter size
 Sows housed in stalls, free-access
 Stalled and group penned sows
stalls, ESF pen, and group pens with
generally had no difference in litter
trickle feeding had a similarity in
size (Harris et al, 2006; Hulbert and
litter size. However, tethered sows
McGlone, 2006; Jansen et al., 2007;
had smaller litter size than stalled and
Karlen et al., 2007; Salak-Johnson et
group penned sows.
al., 2007; Chapinal et al., 2010a).
However, Seguin et al.(2006)
reported higher litter size in group
penned sows than stalled sows
(10.33 ± 0.20 vs. 9.59 ± 0.34
piglets). For group penned sows, the
increase in floor space resulted in
the greater litter size (Salak-Johnson
et al., 2007).
Birth weight
 Birth weights of piglets of sows
housed in stalls did not differ from
 Stalled or group penned sows
those from sows housed in freeproduced piglets with similar
access stalls and group pens with
average birth weight and litter birth
trickle feeding but higher than those
weight (Harris et al., 2006; Hulbert
from sows kept in ESF pens. Girthand McGlone, 2006; Karlen et al.,
tethered sows had piglets with higher
2007; Salak-Johnson et al., 2007;
birth weights than stalled sows.
Chapinal et al., 2010a). However,
Seguin et al. (2006) observed higher
piglet birth weights in group penned
sows than stalled sows (1.58 ± 0.02
vs 1.52 ± 0.03 kg).
 A higher proportion of piglets born
from gilts kept in stalls needed
liquid feed after weaning and drank
more water at 2 d after weaning
(Sorrells et al. 2006)
 Body weight at 35 d of piglets born
from gilts housed in stalls was lower
than those born from gilts kept in
group pens (Sorrells et al. 2006)
 No differences in salivary cortisol
and immune responses of piglets
born from gilts reared in stalls or
group pens (Sorrells et al., 2006)
25
ScholarOne, 375 Greenbrier Drive, Charlottesville, VA, 22901
Professional Animal Scientist-Review copy--Confidential
Page 26 of 50
1
Running Heading: Updated Review of Gestation Sow Housing Literature
Review: Updated scientific evidence on the welfare of gestating sows kept
in different housing systems
John J. McGlone1, PAS, ACAWS
Laboratory of Animal Behavior, Physiology and Welfare2, Department of Animal and
Food Sciences, Texas Tech University, Lubbock, TX 79409-2141
r
Fo
Correspondence:
John J. McGlone, PhD
Professor and Institutional Official
Animal and Food Sciences Department
Texas Tech University
Lubbock, TX 79409-2141
phone: 806-742-2805, ext. 246
fax: 806-742-4003
john.mcglone@ttu.edu
er
Pe
ew
vi
Re
1
2
Correspondence: john.mcglone@ttu.edu.
http://www.depts.ttu.edu/animalwelfare/
1
ScholarOne, 375 Greenbrier Drive, Charlottesville, VA, 22901
Page 27 of 50
Professional Animal Scientist-Review copy--Confidential
2
ABSTRACT
1
Gestation sow housing is a contemporary animal welfare issue with legislative actions in the
3
United States to ban individual gestation sow systems. This review sought to summarize the
4
scientific literature since earlier reviews were published in 2004 and 2005. Seventeen papers
5
comparing effect of housing systems on the welfare of gestating sows were published from
6
2005 to 2012. Stalls/crates, tether housing, and group pens including conventional group
7
pens, loose-house pens, electronic sow feeders (ESF) pens, and hoop barns were reported.
8
The majority of recent findings were similar to the previous conclusions from papers and
9
reviews. Cortisol concentrations were not different between stalled and group penned sows
r
Fo
2
but were higher shortly after mixing in concert with increased sow aggression and skin
11
lesions. The effects of gestation housing system on neutrophil:lymphocyte ratio was still
12
mixed. In the recent literature, stereotypies were greater among stalled sows compared with
13
group penned sows which may reflect a change in sow genetics since the basic housing
14
systems have not changed. Recent papers indicated more sitting/standing inactive, leg and
15
claw problems, and higher farrowing rate (in some studies) among stalled sows. Other studies
16
reported more lying down among group penned sows, and no differences in other
17
reproductive and productivity measures between stalled and group penned sows. Weaning-to-
18
estrus intervals were similar between stalled and group penned sows in contrast to previous
19
reviews that observed a shorter interval among stalled compared with grouped sows. In
20
conclusion, recent papers found similar productivity, physiology health and behavior among
21
individually or grouped sows during gestation.
er
Pe
10
ew
vi
Re
22
23
Key words: Pig, Sow, Gestation housing, Welfare
24
25
2
ScholarOne, 375 Greenbrier Drive, Charlottesville, VA, 22901
Professional Animal Scientist-Review copy--Confidential
Page 28 of 50
3
26
INTRODUCTION
27
28
Gestation sow housing is a social issue in Europe, North America, Australia and New
Zealand. The primary concern is about the welfare of pregnant sows kept in individual
30
gestation crates (also called stalls). The gestation crate does not allow the sow to turn around
31
or make normal postural adjustments. However, the gestation crate is the most common
32
system used in the United States because it minimizes space needs and provides a means to
33
individually feed and care for pregnant sows. Tether housing and crates or stalls for
34
individual sows have been perceived as housing systems that cause poor welfare. Therefore,
35
European countries and several states (FL, CA, AZ, OR, CO, RI, at least) have banned or are
36
phasing out gestation crates. Retailers (grocery and restaurant) have made public statements
37
that they will preferentially buy pork from farms that do not use gestation crates. Still, the
38
industry defends the gestation crate as a viable sow housing system.
er
Pe
Two reviews of the scientific literature about the welfare of pregnant sows were
Re
39
r
Fo
29
published in 2004 and 2005 (McGlone et al., 2004; Rhodes et al., 2005). In the past 8 yr
41
more research has been published that sought to compare the welfare of pregnant sows in
42
individual gestation crates and group pens. The objectives of this review were (a) to
43
summarize the current scientific literature on the welfare of gestation sow housing systems
44
and (b) to determine if the conclusions of the recent literature differ from the conclusions of
45
the previous reviews. Finally, the current status of the welfare of pregnant sows in housing
46
systems is presented based on the newer scientific literature.
ew
vi
40
47
48
SYSTEMS REVIEWED
49
The “control” system by which other systems are compared is the gestation stall (also
50
called crate). This is the control or standard system because at this time, in North America, it
3
ScholarOne, 375 Greenbrier Drive, Charlottesville, VA, 22901
Page 29 of 50
Professional Animal Scientist-Review copy--Confidential
4
is the most common system to keep pregnant sows. Any change away from the crate will
52
incur an economic cost to pork producers and ultimately the consumer. The industry and
53
consumers should know if this seemingly inevitable economic cost is associated with better,
54
the worse or the same sow welfare. While a plethora of systems can be found in which to
55
keep pregnant sows, only a few have generated sufficient scientific literature in which
56
comparisons can be made. These group-housing systems include the (a) group penning with
57
or without individual feeding stalls, and (b) Electronic sow feeders (ESF). Other systems
58
included in the earlier reviews (McGlone et al., 2004; Rhodes et al., 2005) from which there
59
was insufficient new literature were the outdoor system, neck or girth tethers, turn-around
60
stalls, and the Hurnik-Morris system. Because only one paper investigated girth tethers and
61
only two papers examined loose-housing in hoop barns, theses system will be mentioned but
62
not highlighted.
er
Pe
63
r
Fo
51
The scope of this updated literature review is to compare individual with group
keeping systems for pregnant sows. Papers have been published with single systems or in
65
ways that ask specific questions about sow preferences or other questions about sow
66
behavior, physiology or health. These papers were not considered here because they did not
67
compare sow keeping systems.
ew
vi
68
Re
64
Group housing systems are more accepted by some people because as they allow the
69
animals to express social behaviors and sows can turn around. However, group pens also pose
70
welfare problems due to fighting of sows to compete for limited resources and in the
71
establishment of a social hierarchy (Spoolder et al., 2009). Adopting a housing system that is
72
good for the overall welfare of the sows should rely on scientific research results that cover
73
several relevant welfare indicators such as behavior, physiology, health, reproduction, and
74
productivity (CAST, 2009).
75
4
ScholarOne, 375 Greenbrier Drive, Charlottesville, VA, 22901
Professional Animal Scientist-Review copy--Confidential
Page 30 of 50
5
MEASURES OF SOW WELFARE
76
77
Multiple measures of sow welfare should be used in any evaluation of
78
housing/penning systems. For this review, as with past reviews, three types of measures were
79
considered to be direct or indirect measures of sow welfare. These categories include (a)
80
physiology, (b) behavior, and (c) health, reproductive performance and productivity. The
81
categories could be split or merged, however, given that relatively few papers collected all
82
measures.
Measures of physiology focused on the stress hormone cortisol (which may rise
r
Fo
83
during stress) and immune measures (which may increase or decrease during stress; Salak-
85
Johnson and McGlone, 2007). Other measures of physiology like heart and respiratory rates
86
were not often measured in the more recent literature.
87
Pe
84
Measures of behavior include maintenance behaviors (sitting, standing, lying, feeding,
er
drinking) and various forms of oral-nasal-facial (ONF) behaviors, including stereotyped
89
behaviors. One particular type of ONF behavior called “stereotypies” is thought by some to
90
be critical measures of sow welfare and thus was summarized. Because of the difficulty of
91
defining which ONF behaviors are and are not “stereotypies, this review considered all ONF
92
behaviors. However, an alternative views relates to expression of more ONF behaviors as a
93
welfare problem (Damm et al., 2005).
ew
vi
94
Re
88
The category of measures including health, reproduction and productivity is very
95
broad. Health measures that were tied to a putative stress response were included in
96
measures of physiology. Measures of reproductive health and productivity are connected and
97
logically can be in one category. Health also includes bone and foot and leg health and body
98
and vulva lesions.
99
100
METHODS
5
ScholarOne, 375 Greenbrier Drive, Charlottesville, VA, 22901
Page 31 of 50
Professional Animal Scientist-Review copy--Confidential
6
101
The scientific literature was searched electronically. Agricola, Pubmed, and Google
102
Scholar were searched from the period of Jan 1, 2005 through August 1, 2012. Keywords
103
included sow and/or pig, housing or penning. In addition, journals known to contain papers
104
on pig biology were searched including The Journal of Animal Science, The Professional
105
Animal Scientist, Livestock Science, Applied Animal Behaviour Science, Animal, Animal
106
Welfare, Journal of Applied Animal Welfare Science and Journal of Swine Health and
107
Production.
r
Fo
108
109
110
FINDINGS
Results from 17 publications from 2005 to 2012 comparing housing systems for
Pe
gestating sows in a variety of aspects relating to welfare were summarized in each animal
112
welfare category. Methodologies used in the reviewed studies are presented in Table 1.
113
Details of the findings by measure are provided in Tables 2 to 6.
er
111
115
Blood or salivary cortisol concentrations have been used to determine stress among
vi
116
Physiology
Re
114
gestating sows in housing systems. Comparing stall- and group-kept sows, most authors
118
found no differences in cortisol concentrations. The recent literature confirms the earlier
119
reviews that sows in stalls or group pens may equally adapt to the environments by mid
120
gestation. Group-kept sows are often mixed, and this results in fighting and an elevation in
121
cortisol (Jansen et al., 2007). Associations among mixing of grouped sows, aggressive
122
interactions, and cortisol have been investigated in studies dealing with grouping of sows in a
123
single system but not including a comparison of sows in crates and group pens. Therefore,
124
these papers were not included in this review. After the social hierarchy was established, no
125
differences were observed in cortisol concentrations between crated and group penned sows
ew
117
6
ScholarOne, 375 Greenbrier Drive, Charlottesville, VA, 22901
Professional Animal Scientist-Review copy--Confidential
Page 32 of 50
7
126
(Hulbert and McGlone, 2006; Karlen et al., 2007). Cortisol was also elevated when group-
127
penned gestating sows were moved to farrowing crates compared with sows crated during
128
gestation and moved to farrowing crates (Sorrells et al., 2007). The restricted movement and
129
individual housing of the farrowing crate may cause the cortisol rise among previously group
130
housed sows.
For stress-related immune measures, there were inconsistent results. Karlen et al.
132
(2007) found higher neutrophil:lymphocyte ratios measured at 15 wk of gestation among
133
stalled sows compared with group penned sows. One could argue that with advancing
134
pregnancy, the crate may become more cramped and therefore elicit a stress response. In
135
contrast, Hulbert and McGlone (2006) did not find differences in neutrophil:lymphocyte
136
ratios, % neutrophil phagocytosis, and neutrophil chemotaxis measured at mid gestation
137
between individually crated and group housed sows; however, they did not collect samples
138
over time of advancing pregnancy.
r
Fo
131
Some authors argue ONF behavior should include feeding and drinking behaviors
vi
141
Behavior
Re
140
er
Pe
139
(McGlone et al., 2004) while others argue ONF should not include feeding and drinking oral
143
behaviors (Damm et al.(2005). Setting this argument aside, one class of ONF behaviors --
144
the stereotypies directed to objects such as bars and feeders -- were greater among stalled
145
sows compared with group penned sows (Karlen et al., 2007). These recent findings are
146
consistent with the results from the previous reviews.
ew
142
147
Sitting or standing inactive for long periods may indicate poor welfare according to
148
both the previous reviews and the current literature. Lying, in the opposite way, may reflect
149
good welfare as shown if the lying behaviors in group penned sows increased. Hulbert and
150
McGlone (2006) found no difference in behavior for sows kept in stalls or group pens.
7
ScholarOne, 375 Greenbrier Drive, Charlottesville, VA, 22901
Page 33 of 50
Professional Animal Scientist-Review copy--Confidential
8
151
Findings about aggressive behaviors were consistent in the recent literature that
evaluated group housing. Sows show an increase in aggressive interactions after weaning or
153
mixing after breeding (Jansen et al., 2007). Seguin et al. (2006) and Salak-Johnson et al.
154
(2007) investigated the effects of space allowance on lesion scores but not on aggressive
155
behavior; nor did they have a direct comparison with individually-kept sows. Clearly, mixing
156
sows into new social groups will elicit aggression that will resolve itself over time. And
157
smaller spaces can lead to increased lesions among group-housed sows, possibly due to
158
increased aggression with smaller floor space (but this was not directly studied).
159
160
Health, Reproductive Performance & Productivity
Pe
161
r
Fo
152
Body lesions (mostly resulting from fighting especially shortly after mixing) and
locomotory problems have been investigated to determine welfare of sows housed in different
163
systems. Concerning foot and leg problems, results were mixed. Conventional group penned
164
sows and those kept in ESF had more lameness scores and claw lesions than stalled sows as
165
long as there was no bedding (Harris et al., 2006; Anil et al., 2007). In contrast, Karlen et al.
166
(2007) reported more lameness among stalled sows than group penned sows provided deep
167
bedding. However, in an epidemiological study, Ryan et al.(2010), observed similar lameness
168
scores and joint pathologies between sows housed in stalls and loose-housed pens where
169
bedding was provided. Therefore, lameness may be more a function of the absence of
170
bedding rather than housing system.
ew
vi
Re
171
er
162
The previous reviews concluded that weaning-to-estrus interval in gestating sows
172
reared in stalls was shorter than those kept in group pens. However, from the current results,
173
weaning-to-estrus interval and also weaning-to-service interval were not different between
174
stalled and group penned sows (Table 5). In one study, farrowing rates were reduced in group
175
penned sows compared to stalled sows (Kareln et al., 2007). However, Hulbert and McGlone
8
ScholarOne, 375 Greenbrier Drive, Charlottesville, VA, 22901
Professional Animal Scientist-Review copy--Confidential
Page 34 of 50
9
176
(2006) did not find a difference in reproduction between crated and grouped sows. Other
177
reproductive parameters also indicate no difference due to housing. On the whole,
178
reproductive performance was similar for sows in group pens and individual crates in the
179
current literature.
180
Several reports showed no difference in litter size for sows housed in stalls compared
with any type of group pens (Harris et al., 2006; Hulbert and McGlone, 2006; Jansen et al,
182
2007; Karlen et al., 2007; Salak-Johnson et al., 2007; Chapinal et al., 2010a). However, one
183
study found greater litter size among group penned sows compared with individually crated
184
sows (Seguin et al., 2006).
185
r
Fo
181
Sorrells et al.(2006) studied prenatal stress in piglets born from gilts housed in stalls
Pe
or pens. Piglets from individually crated mothers were lighter and required more
187
supplemental feeding than piglets whose mothers were group housed. These findings may not
188
be directly relevant to the welfare of gestating sows but may relate to the welfare in the entire
189
system since the welfare of piglets may be impacted by gestation housing system.
er
186
Re
190
CONCLUSIONS
192
Cortisol concentrations may be used as a physiological sign of stress. Cortisol
ew
vi
191
193
concentrations were higher among group penned sows shortly after mixing (Jansen et al.,
194
2007). However, cortisol concentrations were not different between stalled and group-
195
penned sows in mid gestation. Differences in immune measures were not detected between
196
crated and penned sows. Physiological measures, on the whole, were not different between
197
individually crated and group-kept pregnant gilts and sows. This is consistent with the earlier
198
reviews.
199
Overall ONF behaviors did not differ among sows in group or individual systems. In
200
contrast, stereotypies were higher among stalled sows than group penned sows in one paper
9
ScholarOne, 375 Greenbrier Drive, Charlottesville, VA, 22901
Page 35 of 50
Professional Animal Scientist-Review copy--Confidential
10
201
(Karlen et al., 2007) but not another (Hulbert and McGlone, 2006). Excessive sitting and
202
standing inactive may indicate poor welfare. More sitting and/or standing inactive were
203
reported in stalled sows compared to groped sows (Munsterhjelm et al., 2008; Weng et al.,
204
2009), however, Hulbert and McGlone (2006) did not find this.
205
Aggression commonly occurs in group penned sows particularly shortly after mixing
206
and during feeding. This aggression leads to body or vulva lesions (Jansen et al., 2007; Salak-
207
Johnson et al., 2007;Chapinal et al., 2010a;Chapinal et al., 2010b).
Because the housing systems have not changed, it is possible that the selection for
209
improved reproductive performance or other traits has resulted in sows that express more
210
stereotypies when individually crated. My laboratory has recent unpublished work using the
211
same model as (Hulbert and McGlone (2006) but with a more modern genetic line and in the
212
recent work, sow aggression was greater among group-housed sows than individually crated
213
sows. However, oOne cannot draw this conclusion on the basis of just a few recent papers or
214
work in progress. However, it is an interesting hypothesis that recent intensive genetic
215
selection may have changed the behavior of pregnant sows perhaps with increased
216
aggressive, sitting and ONF behaviors.
er
Pe
vi
Re
217
r
Fo
208
Both recent studies and the earlier reviews found that leg and foot problems had a
ew
218
greater incidence in group penned sows than crated sows (McGlone et al., 2004; Harris et al.,
219
2006; Anil et al., 2007). However, lameness issues may be more a function of bedding (or
220
lack of bedding) than penning system (Karlen et al., 2007).
221
In the recent literature, no differences were observed in weaning-to-estrus interval
222
among stalled versus group-kept sows. Higher farrowing rates were reported among stalled
223
sows compared to group penned sows in one study (Karlen et al., 2007), but this was not
224
found by Hulbert and McGlone (2006). Other reproductive parameters such as ovulation rate,
225
number of embryos, embryo survival, backfat thickness, and body condition scores were
10
ScholarOne, 375 Greenbrier Drive, Charlottesville, VA, 22901
Professional Animal Scientist-Review copy--Confidential
Page 36 of 50
11
226
similar among sows kept in stalls or group pens. Most studies indicated no difference in litter
227
size and birth weight of piglets born from sows housed in stalls or group pens. The majority
228
of the current findings from 2005 to 2012 were not different from those in the previous
229
review. On the whole, few physiological, behavioral or health differences were reported
230
between individually crated and group-kept pregnant gilts and sows.
231
232
Overall, group penning and individual crating of pregnant sows support about the
r
Fo
233
Implications
same level of measurable sow welfare. Grouping has negative consequences such as more
235
body lesions occur due to aggressive behaviors when grouped. Individual crating may cause
236
more stereotyped behaviors of unknown cause or consequence, although this observation was
237
not consistent across studies. A compilation of the recent scientific literature does not draw
238
one to a different conclusion than the literature summarized in 2004 and 2005.
239
Improvements in the welfare of sows are possible in each pregnant sow keeping system.
er
Pe
234
Re
vi
ACKNOWLEDGEMENTS
The authors thank Dr. Kim Guay for technical review of this paper and Wirawat
ew
Chaya for gathering the literature. The author declares no conflicts of interest during the
writing of this review.
LITERATURE CITED
Chapinal, N., J. L. Ruiz de la Torre, A. Cerisuelo, J. Gasa, M. D. Baucells, J. Coma, A. Vidal,
and X. Manteca. 2010a. Evaluation of welfare and productivity in pregnant sows kept in
stalls or in 2 different group housing systems. J. Vet. Behav. 5:82.
Chapinal, N., J. L. Ruiz de la Torre, A. Cerisuelo, J. Gasa, M. D. Baucells, and X. Manteca.
2010b. Aggressive behavior in two different group-housing systems for pregnant sows. J.
Appl. Anim. Welf. Sci. 13:137.
11
ScholarOne, 375 Greenbrier Drive, Charlottesville, VA, 22901
Page 37 of 50
Professional Animal Scientist-Review copy--Confidential
12
Council for Agricultural Science and Technology (CAST). 2009. Scientific Assessment of
the Welfare of Dry Sows Kept in Individual Accommodations. Issue Paper 42. CAST, Ames,
Iowa.
Damm, B. I., B. Forkman, M. Giersing, J. Ladewig, and H. Houe. 2005. Letter to the editor:
An alternative perspective on the compilation of the scientific literature comparing housing
systems for gestating sows using metaanalysis on measures of physiology, behavior,
performance, and health. Prof. Anim. Sci. 21:139.
Harris, M. J., E. A. Pajor, A. D. Sorrells, S. D. Eicher, B. T. Richert, and J. N. MarchantForde. 2006. Effects of stall or small group gestation housing on the production, health and
behavior of gilts. Livest. Sci. 102:171.
r
Fo
Hulbert, L. E. and J. J. McGlone. 2006. Evaluation of drop versus trickle-feeding systems for
crated or group-penned gestating sows. J. Anim. Sci. 84:1004.
Jansen, J., R. N. Kirkwood, A. J. Zanella, and R. J. Tempelman. 2007. Influence of gestation
housing on sow behavior and fertility. J. Swine Health Prod. 15:132.
Pe
Karlen, G. A. M., P. H. Hemsworth, H. W. Gonyou, E. Fabrega, A. D. Strom, and R. J.
Smits. 2007. The welfare of gestating sows in conventional stalls and large groups on deep
litter. Appl. Anim. Behav. Sci. 105:87.
er
Lammers, P. J., M. S. Honeyman, J. W. Mabry, and J. D. Harmon. 2007. Performance of
gestating sows in bedded hoop barns and confinement stalls. J. Anim. Sci. 85:1311.
Re
vi
McGlone, J. J., E. H. von Borrel, J. Deen, A. K. Johnson, D. G. Levis, M. Meunier-Salaun, J.
Morrow, D. Reeves, J. L. Salak-Johnson, and P. L. Sunberg. 2004. Review: Compilation of
the scientific literature comparing housing systems for gestating sows and gilts using
measures of physiology, behavior, performance, and health. Prof. Anim. Sci. 20:105.
ew
Munsterhjelm, C. A. Valros, M. Heinonen, O. Halli, and O. A. T. Peltoniemi. 2008. Housing
during early pregnancy affects fertility and behavior of sows. Reprod. Dom. Anim. 43:584.
Rhodes, R. T., M. C. Appleby, K. Chinn, L. Douglas, L. D. Firkins, K. A. Houpt, C. Irwin, J.
J. McGlone, P. Sundberg, L. Tokach, and R.W. Wills. 2005. A comprehensive review of
housing for pregnant sows. J. Am. Vet. Med. Assoc. 227:1580.
Ryan, W. F., P. B. Lynch, and J. V. O’Doherty. 2010. A survey of bone integrity from cull
sows in Ireland. Irish Vet. J. 63:689.
Salak-Johnson, J . L., S. R. Niekamp, S. L. Rodriguez-Zas, M. Ellis, and S. E. Curtis. 2007.
Space allowance for dry, pregnant sow in pens: Body condition, skin lesions, and
performance. J. Anim. Sci. 85:1758.
Salak-Johnson, J. L., and J. J. McGlone. 2007. Making sense of apparently conflicting data:
stress and immunity in swine and cattle. J. Anim. Sci. 85: Suppl: E81-8.
12
ScholarOne, 375 Greenbrier Drive, Charlottesville, VA, 22901
Professional Animal Scientist-Review copy--Confidential
Page 38 of 50
13
Seguin, M. J. D. Barney, and T. M. Widowski. 2006. Assessment of a group-housing system
for gestating sows: Effects of space allowance and pen size on the incidence of superficial
skin lesions, changes in body condition, and farrowing performance. J. Swine Health Prod.
14:89.
Sorrells, A. D., S. D. Eicher, M. J. Harris, E. A. Pajor, and B. T. Richert. 2007. Periparturient
cortisol, acute phase cytokine, and acute phase protein profiles of gilts housed in groups or
stalls during gestation. J. Anim. Sci. 85:1750.
Sorrells, A. D., S. D. Eicher, K. A. Scott, M. J. Harris, E. A. Pajor, D. C. Lay, Jr., and B. T.
Richert. 2006. Post natal behavioral and physiological responses of piglets from gilts housed
individually or in groups during gestation. J. Anim. Sci. 84:757.
r
Fo
Spoolder, H. A. M., M. J. Geudeke, C. M. C. Van der Peet-Schwering, and N. M. Soede.
2009. Group housing of sows in early pregnancy: A review of success and risk factors.
Livest. Sci. 125:1.
Pe
Anil, S. S., L. Anil., J. Deen, and S. K. Baidoo. 2007. Factors associated with claw lesions in
gestating sows. J. Swine Health Prod. 15:78.
er
van der Staay, F. J., T. Schuurman, M. Hulst, M. Smits, J. Prickaerts, G. Kenis, and S. M.
Korte. 2010. Effects of chronic stress: A comparison between tethered and loose sows.
Physiol. Behav. 100:154.
Re
van Wettere, W. H. E. J., S. J. Pain, P. G. Stott, and P. E. Hughes. 2008. Mixing gilts in early
pregnancy does not affect embryo survival. Anim. Rep. Sci. 104:382.
ew
vi
Weng, R. C., S. A. Edwards, and L. C. Hsia. 2009. Effect of individual, group or ESF
housing in pregnancy and individual or group housing in lactation on sow behavior. AsianAust. J. Anim. Sci. 22:1574.
13
ScholarOne, 375 Greenbrier Drive, Charlottesville, VA, 22901
Page 39 of 50
Professional Animal Scientist-Review copy--Confidential
14
Table 1. Methodologies in reviewed recent literature.
Authors
Animals, parities
Housing
systems
compared
Chapinal et al.
Landrace x Large
Stall (S)
(2010a)
White; parities 1-9 group pen (P)
unprotected
ESF group pen
(ESF)
Fo
rP
Space
allowance
(m2/sow)
1.29 (S)
2.76 (P)
2.32
(ESF)
ee
Bedding
Group sizes,
replicates
No
20 stalled sows, 2
group pens of 10
sows, and an ESF
pen of 20 sows x 3
replicates
rR
Chapinal et al.
(2010b)
Landrace x Large
White; parities 1-9
Group pen (P)
ESF group pen
(ESF)
2.76 (P)
2.32
(ESF)
No
Ryan et al. (2010)
Cull sows in 8-9
parities (studied at
the abattoir)
Stall, loosehouse
NA
Yes
(straw for
loosehouse)
Great
Yorkshire/Large
White x Dutch
Landrace; parities
2-13
Landrace x
Yorkshire
gilts
Tethering (T), 2.2 (L)
loose house (L)
van der Staay et
al.(2010)
Weng et al.(2009)
Stall S)
group pen (P)
ESF group pen
(ESF)
1.41(S)
1.80 (P)
NA (ESF)
2 group pens of 10
sows and an ESF
pen of 20 sows x 3
replicates
4,120 stalled sows
from 5 farms, 2,900
loose-housed sows
from 2 farms (group
size of 100)
59 loose-housed
sows (65-130
sows/pen), 70
tethered sows
ev
No
No
iew
16 stalled sows, 16
group penned sows
(5 sows/pen), 32
ESF sows
14
ScholarOne, 375 Greenbrier Drive, Charlottesville, VA, 22901
Measures
•
•
•
Behavior (week 1-2 and
then every week)
Haptoglobin and PigMAP (week 11-13 and
16-18)
Reproduction
Productivity
Lameness and injury
•
Aggressive behavior
•
•
Lameness score
Joint pathologies
•
•
Plasma cortisol
Weights of adrenal,
pituitary, and spleen
mRNA of PrP and βglobin in brain
Behavior at 8
physiological stages from
estrus to dry
•
•
•
•
Professional Animal Scientist-Review copy--Confidential
Page 40 of 50
15
Munsterhjelm et
al.(2008)
van Wettere et
al.(2008)
Anil et al.(2007)
Jansen et al.(2007)
Yorkshire (64%),
Yorkshire x
Finnish Landrace
(36%), averaged
parities 2.4
Purebred maternal
(Large
White)/terminal
(Duroc)
Gilts and sows;
parities 0-8
Stall, group
pen
Landrace and
Yorkshire; parities
0, 1, 3
Stall, group
pen
Fo
Karlen et al.(2007) Landrace x Large
White; parities ≥1
Stall, group
pen
1.44 (S)
5.10 (P)
rP
Stall, ESF
group pen,
Stall, group
pen
Yes (deep
litter for
pen)
240 stalled sows,
240 group penned
sows (12 pens of 20
sows)
1.36 (S)
2.40 (P)
No
1.20 (S)
NA (ESF)
No
24 stalled sows, 4
group pens of 6 sows
x 3 mixing
treatments
82 stalled sows, 102
group penned sows
(no group size
mentioned)
48 stalled sows, 48
group penned sows
(mixed with other
sows in 4 pens to
form a group of 50)
ee
rR
NA (S)
2.10 (P)
1.26 (S)
2.30 (P)
No
ev
Yes (deep
rice hull
for group
pen)
iew
640 sows in total,
group size of 85
•
•
Behavior (day 27 after
weaning)
Reproduction
Productivity
•
Reproduction
•
Claw lesions measured at
day 110 of gestation
•
Behavior (0-3 d after
mixing)
Lesion scores (+1, -1 d of
mixing)
Salivary cortisol (+1, -1 d
of mixing)
Reproduction
Behavior (week 1, 9 of
gestation)
Salivary cortisol (day 5,
week 9 of gestation)
Hematology and
lymphocyte proliferation
(week 15 of gestation)
Reproduction
Productivity
•
•
•
•
•
•
•
Lammers et
al.(2007)
25% Hamshire x
50% Yorkshire x
25% Landrace;
Stall (S), hoop
barns (H)
1.26 (S)
3.44 (H)
Yes (deep
cornstalks
for hoop
552 stalled sows,
405 group penned in
hoop barns (32
15
ScholarOne, 375 Greenbrier Drive, Charlottesville, VA, 22901
•
•
Page 41 of 50
Professional Animal Scientist-Review copy--Confidential
16
parities 0-2
barns)
Salak-Johnson et
al.(2007)
Sows (PIC) in
parities ≥1
Sorrells et
al.(2007)
Landrace x
Yorkshire gilts
Stall, group
pen
1.29 (S)
1.40,
2.30, and
3.30 (P)
1.32 (S)
2.34 (P)
Harris et al.(2006)
Landrace x
Yorkshire gilts
Stall, group
pen
1.35 (S)
2.40 (P)
Hulbert and
McGlone (2006)
Stall, group
pen
Fo
Camborough 22
gilts
rP
Stall, group
pen
ee
No
16 stalled gilts, 8
group pens of 4 gilts
•
No
16 stalled gilts, 8
group pens of 4 gilts
•
•
No
80 stalled gilts, 80
group penned gilts
ev
iew
Seguin et al.(2006) Yorkshire; parities
2.8
Sorrells et
al.(2006)
Landrace x
Yorkshire gilts
Stall, group
pen
Stall, group
pen
•
•
No
rR
1.33 (S)
1.28 (S)
sows/pen, 2
pens/barn, 2 barns)
51 stalled sows, 166
group penned sows
(group size of 5)
2.00 (S)
2.30,
2.80, and
3.20 (P)
No
1.35 (S)
2.41 (P)
No
98 stalled sows, 9
small group penned
sows (group size of
11-19), 6 large group
penned sows (group
size of 22-31)
16 stalled gilts, 8
group pens of 4 gilts
16
ScholarOne, 375 Greenbrier Drive, Charlottesville, VA, 22901
•
•
•
•
•
•
•
•
•
Productivity
Lesion scores (every 3 d
for the first 2 weeks, then
every week of gestation)
Cortisol , cytokines, and
acute phase proteins
during gestation and after
farrowing
Behavior (week 4, 6, 9,
and 13 of gestation)
Lesion scores (every 2
weeks)
Productivity
Behavior (50-60 d of
gestation)
Cortisol (50-60 d of
gestation)
Hematology and immune
(50-60 d of gestation)
Reproduction
Productivity
Lesion scores (+1,-1 d of
mixing, then every week
for 5 weeks)
Productivity
Behavior (isolation test in
piglets)
Professional Animal Scientist-Review copy--Confidential
Page 42 of 50
17
•
Fo
rP
ee
rR
ev
iew
17
ScholarOne, 375 Greenbrier Drive, Charlottesville, VA, 22901
Cortisol, acute phase
proteins, IgG, and TNF-α
in piglets
Page 43 of 50
Professional Animal Scientist-Review copy--Confidential
18
r
Fo
Table 2. Effects of housing on physiology of gestating sows
Findings from the previous reviews in 2004
Findings from 2005-2012
(McGlone et al., 2004) and Rhodes et
al.(2005)
Cortisol
• Tethered sows had higher plasma
• Circulating cortisol concentrations of
cortisol than loose-housed sows
stalled and group penned sows did not
(van der Staay et al., 2010).
differ. However, the concentrations in
No variation in concentrations of
tethered sows were greater compared
plasma cortisol measured during 50to those in stalled or group penned
60 days of gestation between sows
sows.
housed in crates and group pens
(Hulbert and McGlone, 2006).
• Salivary cortisol concentrations
measured at 1 and 9 weeks of
gestation were not different between
stalled and group penned sows
(Karlen et al., 2007). However,
Jansen et al. (2007) found
significantly higher salivary cortisol
concentrations in group penned
sows the day of relocation compared
to prior moving. In addition, group
penned gilts had higher salivary
cortisol levels shortly after moving
to farrowing crates compared to
stalled gilts (Sorrells et al., 2007).
er
Pe
Re
•
There were no differences in
neutrophil: lymphocyte ratio, %
neutrophil phagocytosis, and
neurophil chemotaxis between
crated and group penned sows
(Hulbert and McGlone, 2006).
Stalled sows had higher neutrophil:
lymphocyte ratio than group penned
sows (Karlen et al., 2007).
No differences in cytokines m RNA
expressions and acute phase proteins
between stalled and group penned
gilts (Sorrells et al., 2007)
•
•
Cardiovascular functions
• There was no information comparing
cardiovascular response of sows in
different housing systems.
•
ew
vi
Immune responses
• Housing did not influence neutrophil:
lymphocyte ratio, immunoglobulin
concentrations, antibody response
against sheep red blood cell antigens,
and natural killer cell activity.
Stalled and group penned sows had
similar heart rates at rest and during
activities (Harris et al., 2006).
18
ScholarOne, 375 Greenbrier Drive, Charlottesville, VA, 22901
Professional Animal Scientist-Review copy--Confidential
Page 44 of 50
19
r
Fo
Table 3. Effects of housing on behavior of gestating sows
Findings from the previous reviews in 2004
Findings from 2005-2012
(McGlone et al., 2004) and Rhodes et
al.(2005)
Oral-nasal-facial (ONF) behaviors and
stereotypies
• Stalled and group penned sows had
• ONF behaviors in stalled and group
no difference in percentage of time
penned sows were similar. Tethered
spent for ONF behaviors (Hulbert
sows might show higher or lower
and McGlone, 2006; Munsterhjelm
ONF behavior depending on
et al., 2008). However, Chapinal et
experimental designs. Sows housed in
al.(2010a) reported that stalled sows
group pens expressed less time spent
spent higher proportion of time for
for stereotypies than those housed in
ONF behaviors than group penned
stalls.
sows. Percentages of time spent for
stereotyped interactions with bars
and feeders were greater in stalled
sows than group penned sows which
was contrast to the percentage of
time spent for stereotyped
interactions with floor (Karlen et al.,
2006).
Locomotory and postural behaviors
• Munsterhjelm et al.(2008) found
• Gilts kept in turn-around stalls had
that stalled sows engaged more in
more standing activity than stalled
standing inactive which was an
gilts. An increase in space allowance
indicator of poor welfare than group
reduced standing and sitting inactive
penned sows. More standing, sitting,
of sows in group pens. Group pen
rooting, and drinking were also
sows changed more postures than
observed in stalled sows than group
stalled sows in farrowing stalls and
penned sows and sows housed in
more restless at parturition.
Electronic Sow Feeding (ESF)
system (Weng et al., 2009). ESF
sows had higher time spent for lying
and less time spent for rooting and
drinking (Weng et al., 2009). This
was consistent with the results of
Chapinal et al. (2010a) indicating
more lying in ESF sows than stalled
sows and group penned sows with
trickle feeding system. Karlen et al.
(2007) reported more lying and less
standing/walking at week 9 of
gestation among stalled sows than
group penned sows. However,
Hulbert and McGlone (2006) did not
find any difference in time budgets
for lying, sitting, and
standing/walking in crated and
group penned sows.
er
Pe
ew
vi
Re
19
ScholarOne, 375 Greenbrier Drive, Charlottesville, VA, 22901
Page 45 of 50
Professional Animal Scientist-Review copy--Confidential
20
Social behaviors
• Agonistic behavior was higher in
group penned sows particularly
during mixing and feeding compared
to stalled and tethered sows.
Increased space allowance resulted in
a decreased agonistic behavior and
social interactions among group
penned sows.
•
r
Fo
er
Pe
Total number of active and passive
aggressive encounters was greater in
group penned sows than stalled
sows and during feeding than during
after mixing (Jansen et al., 2007).
When 2 types of group pens were
compared, ESF penned sows had
higher frequency of aggressive
behavior than conventional group
penned sows (Chapinal et al.,
2010b). Karlen et al.(2007) also
noted the decrease of fighting in
group penned sows at 9 weeks of
gestation compared to that in the
first week. Results revealed by
Hulbert and McGlone (2006)
indicate no difference in agonistic
behavior of crated sows and group
penned sows at 50-60 days of
gestation.
ew
vi
Re
20
ScholarOne, 375 Greenbrier Drive, Charlottesville, VA, 22901
Professional Animal Scientist-Review copy--Confidential
Page 46 of 50
21
r
Fo
Table 4. Effects of housing on health of gestating sows
Findings from the previous reviews in 2004
Findings from 2005-2012
(McGlone et al., 2004) and Rhodes et
al.(2005)
Vulva biting
• Higher numbers of sows developed
• Vulva injuries related to biting were
vulva biting in ESF and group
more frequent in ESF sows than other
penned sows compared to stalled
types of group housing systems.
sows (Chapinal et al., 2010).
There was no prevalence of vulva
However, Harris et al. (2006) did
biting in stalled sows and tethered
not detect difference in vulva lesion
sows.
scores between stalled and group
penned sows.
Body lesion scores
• Group penned sows had higher
• Group penned sows had 3 times
lesion scores than stalled sows after
higher percentage of body lesions
mixing and these scores correlated
caused by fighting compared to
with the total number of agonistic
stalled sows. Provision of roughage
interactions (Jansen et al., 2007).
reduced a risk of developing body
Salak-Johnson et al.(2007) noticed a
lesions in group penned sows.
decrease in space allowance (1.4 vs.
2.3 and 3.3 m2) leading to higher
lesion scores in group penned sows.
However, a study by Seguin et
al.(2006) indicated that there was no
difference in lesion scores of sows
provided space allowance of 2.3,
2.8, and 3.2 m2. The study by
Hulbert and McGlone (2006)
showed no dissimilarity in lesion
scores measured at mid-gestation
between crated sows and group
penned sows.
er
Pe
Locomotory problems
• Incidence of claw lesions was twice
as high among loose-housed sows
than stalled sows and was also more
common than tethered sows. Sows
housed in group pens with deep
bedding were found to have higher
claw lesions than stalled and tethered
sows. Without bedding, lameness
incidence was greater in ESF and
group penned sows compared to
ew
vi
Re
•
Group penned gilts lacking bedding
had more foot and leg problems
investigated at 13 weeks of gestation
than stalled gilts (Harris et al.,
2006). Karlen et al.(2007) reported
higher lameness scores measured at
9 and 13 weeks of gestation among
stalled sows than group penned
sows provided deep bedding.
Comparing ESF with stall housing,
Anil et al.(2007) discovered higher
proportion of sows kept in ESF
developing claw lesions. However,
based on the study of Ryan et
al.(2010), lameness scores and joint
pathology of culled sows previously
21
ScholarOne, 375 Greenbrier Drive, Charlottesville, VA, 22901
Page 47 of 50
Professional Animal Scientist-Review copy--Confidential
22
stalled and free –access stalled sows.
housed in stalls or loose-housed
pens were not different
r
Fo
er
Pe
ew
vi
Re
22
ScholarOne, 375 Greenbrier Drive, Charlottesville, VA, 22901
Professional Animal Scientist-Review copy--Confidential
Page 48 of 50
23
r
Fo
Table 5. Effects of housing on reproductive performance of gestating sows
Findings from the previous reviews in 2004
Findings from 2005-2012
(McGlone et al., 2004) and Rhodes et
al.(2005)
Weaning-to-estrus interval
• Sows housed in stalls had shorter
• Munsterhjelm et al.(2008) observed
weaning-to-estrus interval than those
no difference in weaning-to-estrus
housed in group pens. Group penned
interval between stalled and group
sows had similar estrus detection rate
penned sows. This was consistent
and estrus duration.
with a report by Jansen et al.(2007).
Also, weaning-to-service did not
differ between stalled and group
penned sows (Harris et al., 2006;
Munsterhjelm et al., 2008). In
contrast, Lammers et al.(2007)
noticed shorter weaning-to-estrus
Farrowing rate
interval among stalled sows than
group penned sows.
• farrowing rate in stalled sows was
higher than group penned sows.
Tethered sows had lower farrowing
• Stalled sows had higher farrowing
rate than stalled and group penned
rate than group penned sows (Karlen
sows. There was no difference in
et al., 2007). In contrary, Hulbert
farrowing rates of outdoor reared
and McGlone (2006) found no
sows compared to indoor stalled
variation in farrowing rates of sows
sows. Stalled sows had greater
housed in stall and group pens.
farrowing rate than loose-housed
sows during summer.
er
Pe
Re
•
•
ew
vi
van Wettere et al.(2008) also
detected a similar in ovulation rate,
number of embryos, and embryo
survival in gilts kept in stalls and
group pens.
Body condition scores (BCS)
evaluated before and after entering
gestation did not differ in stalled and
group penned sows (Seguin et al.,
2006). Body weight and backfat
thickness were also not different
between stalled and group penned
sows (Harris et al., 2006; Hulbert
and McGlone, 2006; Chapinal et al.,
2010a). Space allowance positively
affected body weight, BCS, and
backfat thickness (Salak-Johnson et
al., 2007). There was a similarity in
body weight at farrowing and
weaning and backfat thickness at
23
ScholarOne, 375 Greenbrier Drive, Charlottesville, VA, 22901
Page 49 of 50
Professional Animal Scientist-Review copy--Confidential
24
breeding, farrowing, and weaning in
sows housed in crates or group pens
(Hulbert and McGlone, 2006).
r
Fo
er
Pe
ew
vi
Re
24
ScholarOne, 375 Greenbrier Drive, Charlottesville, VA, 22901
Professional Animal Scientist-Review copy--Confidential
Page 50 of 50
25
r
Fo
Table 6. Effects of housing during gestation on productivity of sows and behavior and
physiology of offspring
Findings from the previous reviews in 2004
Findings from 2005-2012
(McGlone et al., 2004) and Rhodes et
al.(2005)
Litter size
• Sows housed in stalls, free-access
• Stalled and group penned sows
stalls, ESF pen, and group pens with
generally had no difference in litter
trickle feeding had a similar litter
size (Harris et al, 2006; Hulbert and
sizes. However, tethered sows had
McGlone, 2006; Jansen et al., 2007;
smaller litter size than stalled and
Karlen et al., 2007; Salak-Johnson et
group penned sows.
al., 2007; Chapinal et al., 2010a).
However, Seguin et al.(2006)
reported higher litter size in group
penned sows than stalled sows
(10.33 ± 0.20 vs. 9.59 ± 0.34
piglets). For group penned sows, the
increase in floor space resulted in
the greater litter size (Salak-Johnson
et al., 2007).
Birth weight
• Birth weights of piglets of sows
housed in stalls did not differ from
• Stalled or group penned sows
produced piglets with similar
those from sows housed in freeaccess stalls and group pens with
average birth weight and litter birth
weight (Harris et al., 2006; Hulbert
trickle feeding but higher than those
and McGlone, 2006; Karlen et al.,
from sows kept in ESF pens. Girth2007; Salak-Johnson et al., 2007;
tethered sows had piglets with higher
Chapinal et al., 2010a). However,
birth weights than stalled sows.
Seguin et al. (2006) observed higher
piglet birth weights in group penned
sows than stalled sows (1.58 ± 0.02
vs 1.52 ± 0.03 kg).
• A higher proportion of piglets born
from gilts kept in stalls needed
liquid feed after weaning and drank
more water at 2 d after weaning
(Sorrells et al. 2006)
• Body weight at 35 d of piglets born
from gilts housed in stalls was lower
than those born from gilts kept in
group pens (Sorrells et al. 2006)
• No differences in salivary cortisol
and immune responses of piglets
born from gilts reared in stalls or
group pens (Sorrells et al., 2006)
er
Pe
ew
vi
Re
25
ScholarOne, 375 Greenbrier Drive, Charlottesville, VA, 22901